Multivariate Analysis Applied to Agglomerated Macrobenthic Data from an Unpolluted Estuary

Marine Environmental Research 87-88 (2013) 112e121

Contents lists available at SciVerse ScienceDirect

Marine Environmental Research

journal homepage: www.elsevier.com/locate/marenvrev

Multivariate analysis applied to agglomerated macrobenthic data from an unpolluted estuary

Anxo Conde a,b,*, Júlio M. Novais a,1, Jorge Domínguez b a IBB-Institute for Biotechnology and Bioengineering, Center for Biological and Chemical Engineering, Instituto Superior Técnico (IST), Av. Rovisco Pais, 1049-001 Lisbon, Portugal b Departamento de Ecoloxía e Bioloxía Animal, Universidade de Vigo, Vigo E-36310, Spain article info abstract

Article history: We agglomerated species into higher taxonomic aggregations and functional groups to analyse envi- Received 20 February 2013 ronmental gradients in an unpolluted estuary. We then applied non-metric Multidimensional Scaling Received in revised form and Redundancy Analysis (RDA) for ordination of the agglomerated data matrices. The correlation be- 15 April 2013 tween the ordinations produced by both methods was generally high. However, the performance of the Accepted 19 April 2013 RDA models depended on the data matrix used to fit the model. As a result, salinity and total nitrogen were only found significant when aggregated data matrices were used rather than species data matrix. Keywords: We used the results to select a RDA model that explained a higher percentage of variance in the species Taxonomic sufficiency Functional groups data set than the parsimonious model. We conclude that the use of aggregated matrices may be Environmental gradients considered complementary to the use of species data to obtain a broader insight into the distribution of Ordination methods macrobenthic assemblages in relation to environmental gradients. Model selection Ó 2013 Elsevier Ltd. All rights reserved. Minho estuary Iberian Peninsula

1. Introduction into consideration the environmental problem in question. Similar conclusions were reached in long term monitoring studies that The distribution of macrobenthic assemblages in relation to propose periodical analysis at species level rather than at lower environmental drivers has been often described at the species level taxonomic resolution, such as the family level (Musco et al., 2011). (Rodrigues et al., 2006; Ysebaert et al., 1998; Zajac and Whitlatch, Agglomeration of species into guilds (usually feeding guilds) has 1982). Agglomeration or aggregation of species at low resolution also been considered (Gaudêncio and Cabral, 2007; García-Arberas levels has also been used to describe environmental gradients. Ac- and Rallo, 2002). The use of functional groups (or functional traits) cording to the concept of taxonomic sufficiency, species data are subdivides the aggregated data at a finer functional level. For aggregated into higher taxonomic levels deemed to be sufficient for instance, the carnivore guild may be further subdivided in accor- the purposes of a study. The term ‘taxonomic sufficiency’ (coined by dance with the motile or sessile characteristic of the species, Ellis, 1985) has been used to identify the effects of pollution on reflecting different ecological niches that are not indicated by the marine communities (e.g. Ferraro and Cole, 1990). The family level ‘carnivore’ grouping alone. Similarly, although species in the same has been proposed as an appropriate level of description for pollu- feeding guild commonly compete for the same food resource, the tion studies (Gómez Gesteira et al., 2003), although higher levels interaction is not necessarily reflected in a taxonomic approach may also be used (Warwick, 1988). However, it has been argued that (Rosenberg, 2001). Improvements in computational methods enable the likelihood of detecting a stressor at higher taxonomic levels will more complex analyses that allow species, biological traits and depend on the severity of the stressor (Ferraro and Cole, 1990), environmental matrices to be considered simultaneously (Dolédec and therefore the taxonomic resolution used in a study must take et al., 1996; Legendre et al., 1997). Analysis and visualization of patterns based on species agglomeration matrices in relation to * Corresponding author. IBB-Institute for Biotechnology and Bioengineering, environmental factors may be accomplished by common ordination Center for Biological and Chemical Engineering, Instituto Superior Técnico (IST), Av. techniques such as Non-metric Dimensional Scaling (NMDS, Terlizzi Rovisco Pais, 1049-001 Lisbon, Portugal. Tel.: þ351 218419124; fax: þ351 et al., 2008) and Redundancy Analysis (RDA, Boström et al., 2006). 218419062. E-mail addresses: [email protected], [email protected] (A. Conde). A few studies have shown that the family level provides an 1 Deceased. adequate description of macrobenthic assemblages along natural

0141-1136/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.marenvres.2013.04.005 A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 113 gradients (De Biasi et al., 2003; Dethier and Schoch, 2006; Moreira et al., 2006). Other authors have recommended using this Wlodarska-Kowalczuk and Kedra, 2007, and references therein). estuary as a pristine reference site for comparison with other metal The family level is a better descriptor than higher taxonomic levels polluted estuaries (Reis et al., 2009). However, Sousa et al. (2008a) or trophic groups for explaining the distribution of the assemblages have identified up to 11 well-established alien species of in- (Dethier and Schoch, 2006), although a lower level of data resolu- vertebrates and fish in the Minho estuary. Among these, the inva- tion may also be useful (Wlodarska-Kowalczuk and Kedra, 2007). sive bivalve Corbicula fluminea (Müller, 1774) has achieved a key Aggregation of species has been underexplored as a means of position in the benthic assemblages because of its high abundance, describing natural gradients (Wlodarska-Kowalczuk and Kedra, biomass and production (Sousa et al., 2008b). 2007). The aim of this study was to assess the usefulness of agglomerated data sets in explaining the distribution of macro- 2.2. Sampling and laboratory procedures benthic assemblages along a natural gradient in an unpolluted estuary. As estuaries may display horizontal, vertical or cross- Sampling was conducted on both banks of the estuary during a sectional spatial gradients, among others (McLusky, 1993), the summer spring tide, between 23 and 26 August 2010. We sampled study was conducted in an area of an estuary between the zones ten sites and assigned them codes in alphabetical order from AM, most affected by marine and freshwater influences. We measured a BM. to KM, excluding IM (Fig. 1). We conducted intertidal sam- number of abiotic variables (redox potential, total nitrogen content pling during low spring tides immediately above the water edge. of the sediments, salinity and others) with the aim of explaining the These sampling surveys included sites along the main axis of the distribution of the macrobenthic assemblages. We compared the estuary, in salt marshes and on an estuarine island (Boega Island, results of applying multivariate analyses to the species dataset and site JM). The length of the sampling area was approximately 13 km, the results of applying the same analyses to agglomerated data extending from the mouth of the estuary to a few kilometres up- grouped at low level of resolution (order and class taxonomic level stream of the village of Vilanova da Cerveira (Fig. 1). and gross feeding guilds), which are rather overlooked in com- We used a standard field probe (WTW 340i) to measure tem- parison with finer aggregation types such as the family taxonomic perature, salinity, oxygen, redox potential in the interstitial water, level (De Biasi et al., 2003) or subdivisions of feeding guilds at a depth of approximate 10 cm in the sediment. We sampled the (Fauchald and Jumars, 1979). The underlying null hypothesis of the top 3 cm of the sediment (making composite sample from three study was that there was no relation between multivariate patterns replicate samples per site) to determine the sediment grain size by in agglomerated data sets and the species data set. the dry sieving method. We defined the grain size fractions as follows: gravel (>2 mm), coarse and medium sand (2e0.250 mm), fi e fi < 2. Material and methods ne sand (0.250 0.063 mm) and nest grains ( 0.063 mm), following the method of Rodrigues et al. (2006) and Silva et al. 2.1. Study site (2006), although these authors considered coarse and medium sand content separately. At each sampling site, we placed samples of the top 3 cm of the sediment in plastic containers and kept them The River Minho is the longest river in the Northwest Iberian Peninsula. The annual average discharge of the River Minho is in a cooler at 4 C for subsequent determination of total organic 13,560 h m3, with a monthly flow average ranging from 100 m3 s 1 carbon and total nitrogen (on dried samples) in an elemental in August to 1000 m3 s 1 in February (deCastro et al., 2006). The analyser (LECO CN2000). estuarine part of the river (Fig. 1) lies between Portugal and Galicia We inserted a corer of inner diameter 95 mm to a depth of 25 cm ¼ 2 (Spain). The Minho estuary has mesotidal features and is partially in the sediment (7 replicates 0.05 m ) to sample infaunal or- mixed, although it tends to be a salt wedge estuary when high ganisms. We sieved all samples through a 1 mm mesh and pre- floods occur (Sousa et al., 2005). Numerous small tributaries drain served the material retained on the mesh in 70% ethanol. We used a into the estuary (Fig. 1). dissecting microscope to sort, count and identify samples of A number of studies have highlighted the low level of anthro- benthic fauna to the lowest possible taxonomic level. pogenic pressure on the Minho estuary (Monteiro et al., 2007; 2.3. Taxa aggregation and categories

We agglomerated the taxa into four aggregation types, each constituted by different categories, as explained below. We first aggregated the taxa into two different taxonomic levels: order and class; the number of categories for both aggregation types depended on the fauna found in each sampling site. We also aggregated the taxa into four categories of trophic guilds: suspension feeders, deposit-feeders (including surface and sub-surface-deposit feeders), carnivores (or predators) and omnivores. This trophic aggregation, with the exception of the omnivore category, follows that used by Chardy and Clavier (1988). We categorized feeding guilds in accordance with Ysebaert et al. (1998), Cummins et al. (1989) and Mancinelli et al. (2005). The feeding guilds are hereinafter referred to as guilds. For the final aggregation, we combined class and guilds in the same data matrix, to produce a mixed data matrix. This approach is based on that used by Pagola-Carte and Saiz-Salinas (2001), although in the present study each taxon occurs twice, as a category of both class and guilds. For example, the species Hediste diversicolor occurs in the polychaete category (class) and the omnivore category (guild). We labelled taxa that did not fitanyof Fig. 1. Map of the Minho estuary showing the location of the sampling sites. the categories within a specific aggregation type as other (e.g. we 114 A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 defined insects as other). We also considered the species data set We used canonical ordination to determine the spatial structure (including some undetermined taxa) in the analysis. of the species assemblages in response to an environmental explanatory matrix. We chose redundancy Analysis (RDA) because 2.4. Statistical analysis it enables transformation of the species data matrix (Legendre and Gallagher, 2001; Borcard et al., 2011) and posterior projection on a We used multivariate analysis to examine the spatial distribution Euclidean space (Legendre and Gallagher, 2001). We constructed and composition of the assemblages within the estuary. We used the RDA triplots by using site scores based on the weighted sum of Vegan package (Oksanen et al., 2011), which is run in the free R species (wa) as a counterbalance or midpoint between the de- environment for statistical computing (R Development Core Team, scriptors used in the analysis and environmental constraints e 2011), for most of the analyses. We calculated a dissimilarity ma- (following Oksanen, 2011). The so-called species environment ’ trix, based on the Euclidean distance of the Hellinger transformed correlation is also provided (Oksanen, 2011). We used Akaike s In- fi raw data, to overcome acute differences in the numbers of information Criteria (AIC) to measure the goodness of t of the RDA dividuals between sampling stations (Legendre and Gallagher, 2001). models (Oksanen et al., 2011). We assessed linear dependencies fl We then applied non-metric Multidimensional Scaling (NMDS) to between environmental variables by computing variance in ation site centroids (7 replicates). NMDS is an ordination technique that factors (Borcard et al., 2011; Oksanen et al., 2011), so that the values displays the distance between the considered objects in accordance of none of the variables was higher than 10, as recommended by with a previously computed dissimilarity matrix (e.g. Legendre and Borcard et al. (2011). We used permutation tests to assess the sig- fi Legendre, 1998). We identified groups of sites by cluster analysis ni cance of the environmental variables and RDA axes of the con- (Legendre and Legendre,1998), in accordance with their similarity in strained ordination. We used forward selection (Blanchet et al., fi species composition. We formed three groups of sites in each cluster 2008) to determine which environmental variables signi cantly analysis by choosing an arbitrary threshold dissimilarity distance explained the distribution of the assemblages. (Clarke and Warwick, 1994). We used complete linkage agglomera- We used Procrustean superimposition (Gower,1971) to compare tive clustering because it is considered an appropriate approach for constrained and unconstrained ordinations between different detecting discontinuities in data (Borcard et al., 2011). species aggregations. Procrustes rotation minimizes the sum of We used non-parametric permutational multivariate analysis of squared residuals between matrices that are compared. Therefore, variance (PERMANOVA, Anderson, 2001, 2005) to test for differ- the optimal superimposition between matrices is used as a metric fi ences between the assemblages in groups identified by cluster of association (Gower, 1971). The statistical signi cance of the dif- analysis, with sites nested within groups of sites. We mainly used ferences highlighted by Procrustean analysis can be assessed by a the PERMANOVA test to assess the convergence of the results ob- permutational procedure (PROTEST; Jackson, 1995). We calculated fi tained with different data sets, not to test for differences on the the corresponding values on the rst two dimensions of the ordi- groups identified a posteriori by cluster analysis, because in such nations to maintain the same dimensionality between the com- cases circularity makes invalid any inference (Clarke and Warwick, parisons (Peres-Neto and Jackson, 2001). We used the PROTEST 1994). We used the Hellinger distance in the analysis. PERMANOVA approach because it has been shown to be more powerful than the enables post-hoc analysis based on uncorrected permutations for Mantel test for this purpose (Peres-Neto and Jackson, 2001). multiple testing. The Monte Carlo asymptotic P-value (Anderson, We assessed and compared the dispersion within each aggre- 2005) is provided as the P-value (P) for the pairwise comparisons. gation type to identify any intrinsic differences in the multivariate We applied the indicator value index (IndVal; Dufrêne and dispersion among aggregation types. Anderson (2001) pointed out Legendre, 1997) to identify the components of the fauna that that the mean distance to the group centroid could be tested in ’ characterized the groups of sites by comparing their abundance and multivariate data in the same way as by a Levene s test for homo- occurrence. In relation to the type of sites under analysis, the geneity of variances in univariate analysis. The comparison used specificity and fidelity of the fauna yield an IndVal value that is here was based on the mean distance for each replicate (7 per site) higher for the most prominent members of the groups; the sig- in relation to the general centroid within an aggregation type prior nificance of the differences in these values is determined by a to Hellinger transformation. Pairwise comparisons can be con- fi permutation test (Dufrêne and Legendre,1997; Borcard et al., 2011). ducted after an overall signi cance test through the calculation ’ fi In IndVal, specificity (designated as Aij, see below) is calculated by of the Tukey s Honest Signi cant Differences (Oksanen, 2011) considering the mean number of individual of species i (or any between aggregations. category in this case) across sites of group j divided by the sum of Values are shown as means standard deviations ( sd). the mean number of individuals of species i over all groups. Fidelity (designated as Bij, see below) is equal to the number of sites in 3. Results cluster j, where species i occurs, divided by the total number of sites in the cluster. Finally, IndVal is calculated as the product between The environmental data obtained at each sampling site are specificity and fidelity: IndVal ¼ Aij *Bij. shown in Table 1. The porewater temperature ranged from

Table 1 Mean values (sd) of the environmental variables measured at each site.

Variable Units AM BM CM DM EM FM GM HM JM KM

Temp C 22.2 2.2 24.0 0.8 23.2 0.5 21.5 0.3 22.3 0.3 22.3 0.3 22.4 1.0 24.0 0.8 24.2 0.6 21.9 0.5 Sal e 17.8 0.8 7.1 0.0 0.8 0.0 10.0 0.0 2.7 0.1 0.7 0.0 6.9 0.1 0.0 0.0 0.0 1 O2 mg l 4.4 0.1 4.3 0.7 4.3 0.0 3.6 0.1 4.1 0.1 3.9 0.1 1.5 0.2 2.7 0.1 4.2 0.0 4.6 0.1 Redox mV 183 1.4 181 0.7 182 0.0 177 4.2 190 2.8 182 0.7 177 1.0 144 2.8 186 0.0 159 0.7 pH e 5.2 0.03 5.8 0.08 6.4 0.06 6.3 0.03 6.1 0.03 6.2 0.01 5.2 0.00 5.8 0.08 6.1 0.06 5.3 0.05 Sand % 0.98 0.51 0.30 0.67 0.59 0.58 0.22 0.39 0.52 0.76 Mud % 0.00 0.02 0.04 0.01 0.02 0.03 0.07 0.01 0.02 0.04 TOC % 0.10 0.12 0.22 0.10 0.16 0.48 0.73 0.10 0.30 0.73 TN % 0.04 0.07 0.09 0.06 0.09 0.10 0.10 0.06 0.07 0.09 Table 2 Mean (sd) density per replicate (0.007 m 2) of the species (or unidenti ﬁed taxa) considered for analysis at each site. Codes for each species, class (Mal, Malacostraca; Oli, Oligochaeta; Pol, Polychaeta; Biv, Bivalvia; Gas, Gastropoda) and feeding guild (DF, Deposit-feeder; SF, Suspension-feeder; O, Omnivorous; P, Predators) are shown. The categories of the order aggregation type are provided in full. Species are shown in decreasing total abundances accounted over the whole set of sampling sites (last column).

Species Code Class Order Guild AM BM CM DM EM FM GM HM JM KM Totals .Cnee l aieEvrnetlRsac 78 21)112 (2013) 87-88 Research Environmental Marine / al. et Conde A. Corophium multisetosum Com Mal Amphipoda DF 2.29 1.8 0.86 0.9 79.86 36.89 68.57 35.93 33 7.66 0.86 0.69 5.57 7.04 1337 Limnodrilus hoffmeisteri Lim Oli Haplotaxida DF 18.14 7.88 1.43 2.94 22.86 22.56 297 Hediste diversicolor Hed Pol Phyllodocida O 1 0.82 7 2.71 7.43 1.9 6.14 2.19 0.86 0.69 0.71 0.75 2.71 1.98 1.14 1.07 0.71 0.49 194 Streblospio shrubsolii Str Pol Spionida DF 9 3.83 9.14 9.15 3.14 4.02 4.71 2.29 0.29 0.76 184 Corbicula ﬂuminea Cof Biv Veneroida SF 0.86 0.69 3.43 1.4 5.71 2.63 7.43 2.7 6.86 1.46 170 Potamopyrgus antipodarum Pot Gas Littorinimorpha DF 1.71 1.6 0.14 0.38 19.29 14.11 2.57 1.51 166 Cyathura carinata Cya Mal Isopoda P 0.29 0.49 2 0.82 1.57 1.72 8.29 2.93 3.71 3.09 0.57 0.79 0.57 0.53 119 Chironomida und. Qui Oth Oth DF 0.57 0.98 0.14 0.38 0.43 0.79 13.14 6.33 100 Psammoryctides barbatus Psa Oli Haplotaxida DF 0.29 0.76 12.71 11.18 91 Insecta und. Ins Oth Oth Ot 0.14 0.38 0.14 0.38 3.43 5.62 26 Oligochaeta und. Oli Oli Oth DF 2.71 3.45 19 Neomysis integer Neo Mal Mysida O 0 0.14 0.38 1.14 2.19 0.49 1.13 0.86 1.21 18 Tubifex tubifex Tub Oli Haplotaxida DF 2.14 4.49 15 Saduriella losadai Sad Mal Isopoda Ot 1 1 0.14 0.38 0.86 0.69 14 Scrobicularia plana Scr Biv Veneroida DF 0.71 0.76 0.43 0.53 8 Gyraulus laevis Gyr Gas Hygrophila DF 1 0.82 7 Cerastoderma edule Cer Biv Veneroida SF 0.29 0.49 0.57 0.53 6 Hydrobia ulvae Hyd Gas Littorinimorpha DF 0.71 0.76 0.14 0.38 6 Crangon crangon Cra Mal Decapoda P 0.43 0.79 3 Gammarus chevreuxi Gam Mal Amphipoda O 0.43 1.13 3 Nematoda und. Nem Oth Oth Ot 0.43 0.53 3

Arenicola marina Are Pol Capitellida DF 0.29 0.49 2 121 Heterotanais oerstedii Het Mal Tanaidacea DF 0.14 0.38 0.14 0.38 2 Lekanesphaera levii Lek Mal Isopoda DF 0.29 0.76 2 Sphaeroma serratum Sph Mal Isopoda DF 0.29 0.49 2 Spiophanes bombyx Spio Pol Spionida DF 0.29 0.49 2 Capitella capitata Cap Pol Capitellida DF 0.14 0.38 1 Scolelepis fuliginosa Sco Pol Spionida DF 0.14 0.38 1 115 116 A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121

3 In total, 70 samples from the ten sampling sites yielded 2798 individuals belonging to 28 different taxa (Table 2). The most 2,5 frequent taxonomic grouping at the order level was Isopoda and at the class taxonomic level, Malacostraca. The most common feeding 2 type was deposit feeders. The ratio of the number of species to the Sp/Or 1,5 number of higher taxa, guilds and mix categories per site is shown Ratio Sp/Cl in Fig. 2. The ratio was maximal for guilds in site KM (2.3 0.4) but 1 Sp/Gu the grand mean of the ratio for all sites was higher for data Sp/Mi aggregated in class (1.5 0.3). The ratio between the number of 0,5 species and the number of classes was higher in the low and mid estuary, whereas the ratio tended to be higher for guilds in the 0 upper estuary. Because the ratio of the number of species to the AM BM CM DM EM FM GM HM JM KM number of categories in the mix aggregation included each species Sites twice, its value was below or equal to 1 for sites along the estuary Fig. 2. Mean values of the ratio (sd) between number of species and number of the (grand mean for all sites 0.7 0.2). following: orders (Sp/Or), classes (Sp/Cl), guilds (Sp/Gu), and mixed categories (Sp/Mi) The ordination of the sites by NMDS yielded similar results for per site. each aggregation type, with a low stress value (maximum 0.03), indicating an excellent fit(Clarke and Warwick, 1994). Cluster 21.5 0.3 Cto24.2 0.6, reflecting summer values. Salinity was analysis identified three groups of sites within each aggregation, maximal close to the mouth of the estuary (site AM) and decreased with the exception of the ordination of guilds, for which four upriver until reaching zero values at the uppermost sampling sta- groups were formed because of the isolated representation of site tions. The south bank tended to higher salinity values relative to the AM (Fig. 3). The stress value for the NMDS based on species (data longitudinal axis (see CM compared with DM and EM compared not shown) was also low (0.03), and the same three clusters of sites with GM sites). Concentrations of dissolved oxygen were similar at as those produced by order, class and mixed aggregation types all sampling sites (approximately 4 mg l 1), except for the mini- were identified. Basically, the clusters of sites reflected their loca- mum values of 2.7 0.1 and 1.5 0.2 mg l 1 recorded at sampling tion along the longitudinal axis of the estuary within the study site sites HM and GM, respectively. The redox potential was positive at (lower, mid and upper location, Figs. 1 and 3). all sites (minimum 144 2.8 in site HM), although the values The results of the PERMANOVA test applied to these groups of tended to be higher in the lower part of the estuary. The pH tended sites (except site AM for the low cluster of sites) were consistent to be low, and the maximum value (6.4 0.06) was measured in with those obtained by NMDS and cluster analysis. The test found sampling site CM. Coarse and medium sand predominated in most significant differences in macrofaunal composition between the of the sampling sites (7 of 10 sites). The organic matter contents three clusters of sites regardless of the aggregation type used for (estimated as total organic carbon) were generally higher in the the analysis. The lowest F value corresponded to the species data uppermost sampling sites, while total nitrogen concentration, set (F2,6 ¼ 8.317, P ¼ 0.001). The pairwise comparisons between measured as a proxy for the quality of the organic matter in the groups were also consistent among the aggregation types; all three sediment, were highest in the mid-estuary sites (FM, GM). groups of sites were found to differ significantly in all cases, and the

Fig. 3. Non-metric Multidimensional Scaling plot displaying the similarity between sites on the basis of the presence of different aggregation types. The groups of sites identified by cluster analysis are also shown (lower, mid and upper estuary). A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 117 highest probability was 0.023 (t ¼ 2.511) for comparison of the . lower and upper clusters of sites within the “order” aggregation. We used the IndVal procedure to identify the faunal categories P Table 2 of the different aggregation types that characterized each of the groups of sites (Table 3). The IndVal procedure revealed different faunal compositions for each cluster of sites, in the same way as identified by NMDS and PERMANOVA. The upper group was best characterized in any aggregation type by a larger number of taxa. Conversely, the middle group was mainly characterized by taxa of the class Malacostraca. Interestingly, the orders Amphipoda and Isopoda were found to be characteristic, but only one amphipod Code Group IndVal Mix species (Corophium multisetosum Stock, 1952) was clearly attribut- DFOliSF MidBiv UpperGas Upper 0.385 0.810 Upper Upper 0.751 0.001 0.646 0.001 0.536 0.001 0.001 0.001 able to the middle group of sites (Tables 2 and 3). The output for the mixed aggregation was identical (by definition) to the joint results 0.001 Pol Lower 0.741 0.001 0.001 O Lower 0.690 0.001 0.001 Mal Mid 0.635 0.001 P < < obtained for the class and guild aggregations. The statistical sig- < nificance (P) in the IndVal analysis was generally higher for the species data set (Table 3). We preceded the search for the best parsimonious RDA model by excluding the mostly highly correlated environmental variables (see variance inflation factor in Section 2.4). We considered a full model, including the set of environmental variables shown in Table 1, except temperature. Temperature is a more appropriate variable for seasonal studies (e.g. Silva et al., 2006) or temporal fl Code Group IndVal gradients (McLusky, 1993). The variance in ation factors (vif) Guild applied to the constrained variables in a full RDA model (based on sites centroid for the species data set) yielded the highest vif value 0.001 O Lower 0.690 0.001 DF Mid 0.385 0.001 SF Upper 0.751 (146) for sediment mud content. We recomputed the full model 0.001 0.001 P < < < without mud as an environmental variable. The remaining envi- < < ronmental variables showed weak linear dependencies. The highest vif value (7.29) in this case corresponded to TN. The resulting full RDA model was significant (P < 0.05) and explained 94% of the observed variance (Table 4). We calculated the parsimonious models centroids of each aggregation type for each site; the resulting triplots are shown in Fig. 4. Most of the parsimonious RDA ordinations revealed one significant environmental variable, usually redox potential (Fig. 4; Code Group IndVal Table 4). This was the case for ordinations based on mixed Class (P ¼ 0.0280) and class (P ¼ 0.025) aggregations and species data set

(P ¼ 0.032) (Fig. 5a; Table 4). The ordination based on order ag- 0.05) within each aggregation type used in the analysis. Codes for the categories of each aggregation type are displayed in 0.001 Pol Lower 0.741 0.001 Mal Mid 0.635 0.001 Oli Upper 0.810 0.0010.001 Gas0.001 Upper 0.536 ﬁ < P < < < < < <

gregation also identi ed redox potential and TN content as signif- P icant factors (P ¼ 0.020 and P ¼ 0.045, respectively). By contrast, the RDA ordination based on guilds identified salinity as a significant factor (P ¼ 0.008; Fig. 4), although this environmental variable was not significant in the remaining ordinations. The model based on the order aggregation explained the highest percentage of variance (48%) and also showed the highest specieseenvironment correla- GroupLower IndVal 0.742 Mid 0.350 0.046 Biv Upper 0.646 tion (0.86), while the ordination based on guilds yielded the best fit (AIC ¼19.85; Table 4). The grouping of sites identified by cluster analysis was still recognizable in the RDA ordination, as shown in Fig. 5a for the species data set. The clustering of sites in RDA by the different aggregation types was similar to that found by MDS (Figs. 3 and 4). The correlations between the categories within each Order Order aggregation type in the RDA triplots were similar to those yielded cant categories submitted to the IndVal test ( by IndVald analysis (Table 3). Finally, the RDA model selected for the species data set (Fig. 5b) included the set of environmental P variables identified by the parsimonious models as exerting significant effects. The model summarized the general spatial distribution of sites in relation to the environmental variables. Thus, the redox potential tended to be higher for the lower and mid group of sites, salinity was higher for the lower group and TN was higher for the mid group (Figs. 4 and 5). The selected model for species data set included the latter environmental variables. This model improved the quantity of variance explained (53%) and the goodness-of-fit (AIC ¼5.14) relative to the parsimonious model Species CodeStr Group IndVal Lower 0.741 0.002 Spionida Hed Lower 0.717 0.002 Phyllodocida Lower 0.717 Scr Lower 0.333 0.002 Amphipoda Mid 0.749 CerHydCom Lower LowerCof Mid 0.190Lim 0.173Pot Upper 0.059Qui Upper 0.748 0.059 0.812Psa Upper Isopoda 0.667Oli Haplotaxida Upper 0.002 0.002 0.583Gyr Upper 0.002 Veneroida 0.442Ins Upper Littorinimorpha 0.002 0.381 Upper Upper Hygrophila 0.002 0.286 Upper Upper 0.810 0.238 0.002 Upper 0.493 0.009 0.234 Upper 0.041 0.646 0.041 0.238 0.02 for the species data set (Table 4), although it included a non- Table 3 Indicator values (IndVald) for the signi fi 118 A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121

Table 4 Summary of the RDA models for each aggregation type. The significant environmental variables are shown, jointly with the explained variance, the correlation between the biotic and environmental variables (I or II first axes), the goodness-of-fit of the models (AIC), the F and significance (P) of each model. DF ¼ degrees of freedom (model, residuals). TN ¼ Total nitrogen; Sal ¼ Salinity.

Category Type of model Environmental Explained Bioticeenv AIC DF (mod, F value Signiﬁcance P tested variables variance (%) correlation value res)

Species Full All 94 I-0.99/II-0.99 18.03 7, 2 4.651 0.002 Species Parsimonious Redox 23 I-0.84 4.1 1, 8 2.332 0.028 Class Parsimonious Redox 31 I-0.83 8.32 1, 8 3.522 0.025 Order Parsimonious Redox þ TN 48 I-0.86/II-0.77 7.33 2, 7 3.219 0.003 Guild Parsimonious Sal 34 I-0.77 19.85 1, 8 4.078 0.012 Mix Parsimonious Redox 25 I-0.81 11.66 1, 8 2.708 0.028 Species Chosen Redox þ Sal þ TN 53 I-0.88/II-0.84 5.14 3, 6 2.278 0.009

significant third constrained axis (variance explained ¼ 3.5%; 4. Discussion F1,6 ¼ 0.697; P ¼ 0.640). The degree of multivariate correlation between the ordinations We have shown that multivariate analysis of agglomerated data obtained by both MDS and RDA ordination methods are shown in grouped at a low level of resolution (at the order and class taxo- Table 5. The MDS ordinations were more consistent than the RDA nomic level and at gross feeding guilds categories) provide similar ordinations. The lowest value corresponded to the comparison results and high degree of correlation and consistency with those between class and guilds aggregation types (m2 ¼ 0.786), although obtained using the species data set. The mixed approach (after the degree of correlation was statistically significant (P ¼ 0.002), as merging the class and guild categories in a single matrix) also in the entire set of comparisons (P ¼ 0.001 in the remaining cases). proved to be useful for describing the distribution patterns of Examination of the correlation between the RDA ordinations intertidal macrobenthic assemblages in the Minho estuary. The use revealed the weakest correlation between the class and guilds ag- of agglomerated data in RDA models also identified as significant a gregations (m2 ¼ 0.734, P ¼ 0.005) and between species and guilds number of environmental variables that would be disregarded by aggregations (m2 ¼ 0.756, P ¼ 0.002). Interestingly, the correlation considering only the species data. Thus, although agglomerated between species and class matrices was higher (m2 ¼ 0.991, data may used to describe patterns of macrobenthic assemblages, P ¼ 0.001) than that between species and order aggregations they were also found to be complementary to the species data set (m2 ¼ 0.967, P ¼ 0.001) among the RDA ordinations. for selecting better RDA models. The test of multivariate homogeneity of group dispersions There is clear evidence that the results of the analysis of ag- detected significant differences in the overall variability across glomerations of macrobenthic species were consistent with pre- aggregation types (F4,345 ¼ 40.715, P < 0.001; Fig. 6). Pairwise vious findings for these species in the Minho estuary. The observed comparisons identified significant differences between aggregation distribution of the intertidal macrofauna is comparable to that types, except between species and order (Padj ¼ 0.964) and be- described by Sousa et al. (2008a) for subtidal macrobenthic as- tween order and class aggregations (Padj ¼ 0.085). The average semblages in similar sampling sites in the Minho estuary. These distance to the aggregation centroid was minimal for the guilds authors also found that salinity was an important environmental (0.449) and maximal for the species data set (0.782). driver explaining the spatial distribution of the assemblages, and

Fig. 4. Triplot graphs of the RDA parsimonious model. Note the different environmental variables found to be signiﬁcant in relation to the aggregation type considered. A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 119

Fig. 5. Triplot graphs of RDA models for species with groups of sites identified by cluster analysis (lower, mid and upper): a) parsimonious model b) selected model. Species codes are shown in Table 2. that grain size and organic matter content were important envi- (Table 4). Muniz and Pires-Vanin (2005) and Olsgard et al. (1997) ronmental constraints within clusters of sites. Mazé et al. (1993) also found different results for the environmental variables in characterized the distribution of the macrobenthic assemblages different aggregation types, although they did not consider these as a sharp transition from marine to freshwater communities in the differences as major findings. Minho estuary. The present results are also consistent with those of The use of taxonomic sufficiency at a family level has been the latter study, although these authors had not yet described recommended for studies involving pollution monitoring (Gómez C. fluminea as part of the macrobenthic assemblages at the upper Gesteira et al., 2003) or identification of natural spatial patterns sampling sites, where species of the class Malacostraca predomi- (Dethier and Schoch, 2006). In some cases, the effect of pollution is nated. The current presence of C. fluminea in the upper reaches has portrayed at lower taxonomic resolution, at the level of order clarified the transition from marine to freshwater assemblages in (Ferraro and Cole, 1990) or even phylum (Warwick, 1988). In both of the Minho estuary (Sousa et al., 2008a). the latter studies, the authors proposed that an environmental The macrobenthic assemblages were grouped into three clusters perturbation is reflected at higher taxonomic levels as the severity in relation to their distribution on the longitudinal axis of the of the perturbation increases. This may be the case in the unpol- Minho estuary (except for the guilds, which did not include site AM luted Minho estuary, in which a strong natural gradient within a in the lower group of sites). Redox potential was the most signifi- distance of a few kilometres ( Fig. 1) allows identification of the cant environmental factor, and it differentiated the lower and mid cluster of sites from the upper group (Figs. 4 and 5; Table 4). Redox potential values are associated with the organic matter content, grain size and chemical forms of nutrients in estuarine sediments (e.g. Mucha and Costa, 1999). Salinity was also identified as a longitudinal environmental descriptor, and it differentiated the lower group from the mid and upper clusters of sites (Fig. 4). The distribution of macrobenthic assemblages in estuaries is largely explained by the salinity gradient (Ysebaert et al., 1998). Moreover, the joint influence of salinity and redox potential as important environmental variables in relation to the distribution of macrobenthic assemblages has been reported previously (Rodrigues et al., 2011). Total nitrogen was associated with the mid-estuary group of sites, which probably indicates some difference in the nutrient quality of the sediment (Rice and Rhoads, 1989). Interestingly, these important environmental factors were found by considering different species aggregation matrices as response variables

Table 5 Degree of correlation between the ordinations obtained by MDS and the RDA parsimonious models. All the correlations between ordinations were statistically signiﬁcant (P < 0.05).

MDS Species Order Class Guild RDA Species Order Class Guild Fig. 6. Boxplot of the multivariate dispersion of the different aggregation types used in Order 0.974 Order 0.967 relation to their general centroid. Horizontal lines within each box show the median Class 0.945 0.978 Class 0.991 0.945 plus the 25th and 75th percentiles at the box ends. Vertical dashed lines show either Guild 0.869 0.815 0.786 Guild 0.756 0.783 0.734 the maximum and minimum value or 1.5 times the interquartile range; individual Mix 0.977 0.976 0.971 0.899 Mix 0.986 0.940 0.993 0.784 points are outliers. 120 A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 distribution of estuarine assemblages from the marine to the 5. Conclusions freshwater edges of the estuary, even with low resolution agglomerated data. Similarly, the nearby poikilohaline Mondego Ordination of the sampling sites by constrained and uncon- estuary characterized by a strong seasonal saline fluctuation was strained multivariate methods provided similar results for the adequately described with regard to the distribution of macro- different levels of species aggregations. The degree of correlation in benthic assemblages from the lower to the upper estuary at the relation to the species ordination was generally high in all cases, but taxonomic resolution of order (Chainho et al., 2007). By contrast, it was weaker for guilds, especially in the RDA ordination. The use the order level was considered a poor habitat descriptor in a rela- of species data sets aggregated into groups of lower taxonomic tive small Mediterranean estuary with a sharp community transi- resolution or functional groups may replace macrobenthic estua- tion (De Biasi et al., 2003). It would be interesting to test rine species data sets in environmental studies. Moreover, the experimentally whether the degree of correlation of the species simultaneous use of various aggregation matrices to explain the data set with aggregation types of lower resolution (such as class or distribution of estuarine macrobenthic assemblages appears to phylum, characterized by a few categories) is higher at higher levels provide complementary information and broader ecological insight of perturbation. This may include coarse aggregations other than than the use of species data alone to describe the assemblages. taxonomic levels, such as guilds, which showed an acceptable Therefore, agglomeration of species into coarse taxonomic or degree of correlation to the ordination based on species (Table 5). functional groups may be used in combination with (rather than The mix aggregation approach appears valid for analysing the instead of) species data sets, for a more comprehensive analysis of distribution and composition of macrobenthic assemblages. In the the distribution of macrobenthic assemblages in relation to envi- present case, the class and guilds matrices were merged to pro- ronmental gradients. duce a matrix of 11 categories, in contrast to the 30-variable- matrix resulting from the combination of 5 (classes) and 6 (guilds) Acknowledgements categories. Each individual is represented twice in the mixed matrix, although under different categories. However, the cate- We are grateful to Dr. Cristóbal for field and lab assistance. We gories may overlap, as was the case for the class Malacostraca and also thank Andrea Tato for laboratory assistance. We are indebted deposit-feeders (Fig. 4, Table 3). This may be complementary to the Galician Minho Fishermen’s Association, especially their rather than redundant information, given that both types of ag- chairman, Mr. Samuel Martínez, for providing a boat to sample the gregation are a priori unrelated because, unlike taxonomic ag- least accessible estuarine sites. The first author was funded by the gregations, functional groups characterize ecosystem processes Portuguese Foundation for Science and Technology (FCT; SFRH/BD/ (Odum, 1969; Rosenberg, 2001). However, this may not always 48928/2008). occur, e.g. in the class Bivalvia, which is largely dominated by suspension-feeding habits or when feeding groups overlap to a References large extent with polychaete families (Fauchald and Jumars, 1979). It would be particularly interesting to assess the usefulness of the Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of mixed aggregation when individual types of aggregation are found variance. Austral Ecology 26, 32e46. Anderson, M.J., 2005. PERMANOVA: a FORTRAN Computer Program for Permuta- to be poor surrogates. Low ratios of the number of species to tional Multivariate Analysis of Variance. Department of Statistics, University of aggregated data sets (as in Fig. 2) usually indicate an adequate Auckland, New Zealand. correlation between similarity matrices (Wlodarska-Kowalczuk Blanchet, F.G., Legendre, P., Borcard, D., 2008. Forward selection of explanatory variables. Ecology 89, 2623e2632. and Kedra, 2007). As shown by the latter authors, higher ratios Borcard, D., Gillet, F., Legendre, P., 2011. Numerical Ecology with R. Springer, New are sometimes associated with poorer correlations of similarity York. between the matrices based on species and matrices constructed Boström, C., O’Brien, K., Roos, C., Ekebom, J., 2006. Environmental variables on aggregated data. Because each species is considered twice in explaining structural and functional diversity of seagrass macrofauna in an archipelago landscape. Journal of Experimental Marine Biology and Ecology the mixed approach, this seems a good strategy for lowering the 335, 52e73. ratio between the species data set and data aggregated at lower Chainho, P., Lane, M.F., Chaves, M.L., Costa, J.L., Costa, M.J., Dauer, D.M., 2007. fi resolution levels, such as the family level, which usually provides Taxonomic suf ciency as a useful tool for typology in a poikilohaline estuary. Hydrobiologia 587, 63e78. satisfactory correlations (Wlodarska-Kowalczuk and Kedra, 2007 Chardy, P., Clavier, J., 1988. Biomass and trophic structure of the macrobenthos in and references therein). This hypothesis needs to be specifically the south-west lagoon of New Caledonia. Marine Biology 99, 195e202. tested in further studies. Regarding the merging of data sets, the Clarke, K.R., Warwick, R.M., 1994. Change in Marine Communities: an Approach to Statistical Analysis and Interpretation. Plymouth Marine Laboratory, Plymouth. mixed ordinations (Figs. 3 and 4) appear to be driven by the subset Cummins, K.W., Wilzbach, M.A., Gates, D.M., Perry, J.B., Taliaferro, W.B., 1989. of data belonging to the most dispersed aggregation, (class ag- Shredders and Riparian vegetation. BioScience 39, 24e30. gregation, Fig. 6). De Biasi, A.M., Bianchi, C.N., Morri, C., 2003. Analysis of macrobenthic communities fi at different taxonomic levels: an example from an estuarine environment in The RDA ordinations probably identi ed different environ- the Ligurian Sea (NW Mediterranean). Estuarine Coastal and Shelf Science 58, mental variables as being important (Fig. 4; Table 4)becauseof 99e106. the different multivariate dispersion within each aggregation deCastro, M., Alvarez, I., Varela, M., Prego, R., Gómez-Gesteira, M., 2006. Miño River dams discharge on neighbor Galician Rias Baixas (NW Iberian Peninsula): hy- type (Fig. 6). The multivariate dispersion of the data sets drological, chemical and biological changes in water column. Estuarine Coastal decreased at lower levels of resolution (Fig. 6), as reported by and Shelf Science 70, 52e62. Terlizzi et al. (2008) and Musco et al. (2011) for the class taxo- Dethier, M.N., Schoch, G.C., 2006. Taxonomic sufficiency in distinguishing natural nomic level. Given that response variables are constrained by spatial patterns on an estuarine shoreline. Marine Ecology Progress Series 306, 41e49. linear combinations of the environmental variables, so that their Dolédec, S., Chessel, D., ter Braak, C.J.F., Champely, S., 1996. Matching species traits dispersion in RDA analysis is maximized (Legendre and Legendre, to environmental variables: a new three-table ordination method. Environ- e 1998), the different nature of the aggregated data used may mental and Ecological Statistics 3, 143 166. fi Dufrêne, M., Legendre, P., 1997. Species assemblages and indicator species: the need explain why different environmental variables are identi ed as for a flexible asymmetrical approach. Ecological Monographs 67, 345e366. being significant. For instance, the agglomeration of species in Ellis, D., 1985. Taxonomic sufficiency in pollution assessment. Marine Pollution guilds allowed identification of salinity as significant because the Bulletin 16, 459. Ferraro, S.P., Cole, F.A., 1990. Taxonomic level and sample size sufficient for spread of this aggregation type better matched the dispersion of assessing pollution impacts on the Southern California Bight macrobenthos. salinity. Marine Ecology Progress Series 67, 251e262. A. Conde et al. / Marine Environmental Research 87-88 (2013) 112e121 121

Fauchald, K., Jumars, P.A., 1979. The diet of worms: a study of polychaete feeding along an established pollution gradient. Marine Ecology Progress Series 149, guilds. Oceanography and Marine Biology: an Annual Review 17, 193e284. 173e181. Gaudêncio, M.J., Cabral, H.N., 2007. Trophic structure of macrobenthos in the Tagus, Pagola-Carte, S., Saiz-Salinas, J.I., 2001. Changes in the sublittoral faunal biomass 511 estuary and adjacent coastal shelf. Hydrobiologia 587, 241e251. induced by the discharge of a polluted river along the adjacent rocky coast (N. García-Arberas, L., Rallo, A., 2002. The intertidal soft-bottom infaunal macrobenthos Spain). Marine Ecology Progress Series 212, 13e27. in three Basque estuaries (Gulf of Biscay): a feeding guild approach. Hydro- Peres-Neto, P.R., Jackson, D.A., 2001. How well do multivariate data sets match? The biologia 475/476, 457e468. advantages of a Procrustean superimposition approach over the Mantel test. Gómez Gesteira, J.L., Dauvin, J.C., Salvande Fraga, M., 2003. Taxonomic level for Oecologia 12, 169e178. assessing oil spill effects on soft-bottom sublittoral benthic communities. Ma- R Development Core Team, 2011. R: a Language and Environment for Statistical rine Pollution Bulletin 46, 562e572. Computing. R Foundation for Statistical Computing, Vienna, Austria. http:// Gower, J.C., 1971. Statistical methods of comparing different multivariate analyses of www.R-project.org. the same data. In: Hodson, F.R., Kendall, D.G., Tautu, P. (Eds.), Mathematics in Reis, P.A., Antunes, J.C., Marisa, C., Almeida, R., 2009. Metal levels in sediments from the Archaeological and Historical Sciences. Edinburgh University Press, Edin- the Minho estuary salt marsh: a metal clean area? Environmental Monitoring burgh, pp. 138e149. and Assessment 159, 191e205. Jackson, D.A., 1995. PROTEST: a Procrustean randomization test of community Rice, D.L., Rhoads, D.C., 1989. Early diagenesis of organic matter and the nutritional environment concordance. Écoscience 2, 297e303. value of sediment. In: Lopez, G.R., Taghon, G.L., Levinton, J.S. (Eds.), Ecology of Legendre, P., Gallagher, E.D., 2001. Ecologically meaningful transformations for Marine Deposit Feeders. Springer-Verlag, New York, pp. 59e97. ordination of species data. Oecologia 129, 271e280. Rodrigues, A.M., Quintino, V., Sampaio, L., Freitas, R., Neves, R., 2011. Benthic Legendre, P., Galzin, R., Harmelin-Vivien, M.L., 1997. Relating behavior to habitat: biodiversity patterns in Ria de Aveiro, Western Portugal: environmental bio- solutions to the fourth-corner problem. Ecology 78, 547e562. logical relationships. Estuarine Coastal and Shelf Science 95, 338e348. Legendre, P., Legendre, L., 1998. Numerical Ecology. Elsevier, Amsterdam. Rodrigues, A.M., Meireles, S., Pereira, T., Gama, A., Quintino, V., 2006. Spatial pat- Mancinelli, G., Sabetta, L., Basset, A., 2005. Short-term patch dynamics of macro- terns of benthic macroinvertebrates in intertidal areas of a Southern European invertebrate colonization on decaying reed detritus in a Mediterranean lagoon estuary: the Tagus, Portugal. Hydrobiologia 555, 99e113. (Lake Alimini Grande, Apulia, SE Italy). Marine Biology 148, 271e283. Rosenberg, R., 2001. Marine benthic faunal successional stages and related sedi- Mazé, R.A., Lastra, M., Mora, J., 1993. Macrozoobentos del estuário del Miño (NO de mentary activity. Scientia Marina 65, 107e119. España). Publicaciones Especiales del Instituto Español de Oceanografía 11, Silva, G., Costa, J.L., Raposo de Almeida, P., Costa, M.J., 2006. Structure and dynamics 283e290. of a benthic invertebrate community in an intertidal area of the Tagus estuary, McLusky, D.S., 1993. Marine and estuarine gradients e an overview. Netherlands Western Portugal: a six year data series. Hydrobiologia 555, 115e128. Journal of Aquatic Ecology 27, 489e493. Sousa, R., Guilhermino, L., Antunes, C., 2005. Molluscan fauna in the freshwater tidal Monteiro, M., Quintaneiro, C., Nogueira, A.J.A., Morgado, F., Soares, A.M.V.M., area of the River Minho estuary, NW of Iberian Peninsula. Annales de Limno- Guilhermino, L., 2007. Impact of chemical exposure on the fish Pomatoschistus logie International Journal of Limnology 41, 141e147. microps Kroyer (1838) in estuaries of the Portuguese northwest coast. Che- Sousa, R., Dias, S., Freitas, V., Antunes, C., 2008a. Subtidal macrozoobenthic as- mosphere 66, 514e522. semblages along the River Minho estuarine gradient (NW of Iberian Peninsula). Moreira, S.M., Moreira-Santos, M., Guilhermino, L., Ribeiro, R., 2006. An in situ Aquatic Conservation: Marine and Freshwater Ecosystems 18, 1063e1077. postexposure feeding assay with Carcinus maenas for estuarine sediment Sousa, R., Rufino, M., Gaspar, M., Antunes, C., Guilhermino, L., 2008b. Abiotic im- overlying, water toxicity evaluations. Environmental Pollution 139, 318e329. pacts on spatial and temporal distribution of Corbicula fluminea (Müller, 1774) Mucha, A.P., Costa, M.H., 1999. Macrozoobenthic community structure in two Por- in the River Minho Estuary, Portugal. Aquatic Conservation: Marine and tuguese estuaries: relationship with organic enrichment and nutrient gradi- Freshwater Ecosystems 18, 98e110. ents. Acta Oecologica 20, 363e376. Terlizzi, A., Anderson, M.J., Bevilacqua, S., Fraschetti, S., Wlodarska-Kowalczuk, M., Muniz, P., Pires-Vanin, A.M.S., 2005. More about taxonomic sufficiency: a case study Ellingsen, K.E., 2008. Beta diversity and taxonomic sufficiency: do higher-level using polychaetes communities in a subtropical bay moderately affected by taxa reflect heterogeneity in species composition? Diversity and Distributions urban sewage. Ocean Science Journal 40, 127e143. 15, 450e458. Musco, L., Mikac, B., Tataranni, M., Giangrande, A., Terlizzi, A., 2011. The use of Warwick, R.M., 1988. Analysis of community attributes of the macrobenthos of coarser taxonomy in the detection of long-term changes in polychaete assem- Frierfjord/Langesundfjord at taxonomic levels higher than species. Marine blages. Marine Environmental Research 71, 131e138. Ecology Progress Series 46, 167e170. Odum, E.O., 1969. The strategy of ecosystem development. Science 164, 262e270. Wlodarska-Kowalczuk, M., Kedra, M., 2007. Surrogacy in natural patterns of benthic Oksanen, J., 2011. Multivariate Analysis of Ecological Communities in R: Vegan distribution and diversity: selected taxa versus lower taxonomic resolution. Tutorial. URL: http://cc.oulu.fi/wjarioksa/opetus/metodi/vegantutor.pdf. Marine Ecology Progress Series 351, 53e63. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., O’Hara, R.B., Simpson, G.L., Ysebaert, T., Meire, P., Coosen, J., Essink, K., 1998. Zonation of intertidal, macro- Solymos, P., Stevens, M.H.H., Wagner, H., 2011. Vegan: Community Ecology benthos in the estuaries of Schelde and Ems. Aquatic Ecology 32, 53e71. Package. http://CRAN.R-project.org/package¼vegan. Zajac, R.N., Whitlatch, R.B., 1982. Responses of estuarine infauna to disturbance. I. Olsgard, F., Somerfield, P.J., Carr, M.R., 1997. Relationships between taxonomic Spatial and temporal variation of initial recolonization. Marine Ecology Progress resolution and data transformations in analyses of macrobenthic community Series 10, 1e14.